Migration synchronisation
We recently migrated from a Progress 7 database on a VMS platform to a brand spanking new Progress 9 database on a Tru64 Unix platform.
During this period we had to ensure data entered into the old system was then sync'd to the new database.
This was done by outputing all updated records to text files (using database write triggers and certain audit tables) on the VMS box and FTPing them to the Tru64 box.
A progress program was then written to take these text files and process them appropriately before applying the changes to the new databases.
Our major problem was that the two databases bore no resemblance schema-wise whatsoever.
The import of the text files to the new system was stopped every night and restarted via cron.
The whole system worked perfectly (or very nearly) for around 18 months while the new database application was developed. To further complicate things, the sync process was tasked with keeping live, UAT, test and development databases in sync with the old platform.
Also, the guy who wrote and managed this whole system says he's available in 3 weeks if you're desperate (it's not me).
Duncan
P.S. Make sure you don't need Latin-1 to Unicode conversions in between the databases as that was a RIGHT pain in the backside.
pcc9011 said:
I am new to this group and very new to progress. I am an Oracle DBA thrown onto a progress project. I need some help/Ideas. Here is the problem. Were a running progress version 7.3 on AIX in production right now and are movein to 9.x on Solaris. So far all is good data migration etc.. The problem is I have to find some way to replicate data from the Soloris to The AIX in case we need to rollback. I am tasked to find a way to keep these DB's in "sync" when we rollout, so if we need to we can jump right back to the AIX.
AI- wont work because of version/platform difference right?
REP Trigs- 7.3C does not support the RAW datatype and I have no clue how to code them..
Dynamic Export takes to long
Anyone have any tricks/Ideas?
